Dirk - there is a vixen setting that causes the intensities to be displayed (and entered) as integers between 0 and 255, rather than between 0 and 100. And I think (but am not sure) that when you specify a ramp, that the intermediate interpolated values are actually integers between 0 and 255.
As I see it, the whole reason for more dimming levels is to enable dimming curves to be implemented in a meaningful way. Without the dimming curves the additional levels are useful only for very slow ramps, as far as I can determine.
EDIT - it has always been my unwritten assumption that dimming curves and more dimming levels go hand-in-hand, and that one without the other is not particularly useful. My apologies for not making this more explicit earlier in the discussion.
Yes! And the purpose of more dimming levels is about linearity... not more levels of control.
For serial driven controllers I think it makes more sense to implement the correction (eg gamma correction) in the controllers because the bottleneck is not in the controller speed... its in the 'pipe'.
I suspect the scenario you mention Phil is what the RGB LED guys are also doing as applying the same voltage to a RED, BLUE, WHITE or GREEN LED will produce different perceived light intensity - beside the physics of the LED itself, our eyes also perceive them differently.
Suggestion:
I don't have my light meter with me --- but from an engineering perspective -- take a light string a measure the intensity at different PWM (assuming AC) levels - they won't be linear - as Phil stated above.
Use those values as an index into a translation array with values that produce the desired curve.
In Vixen 3 - this translation is super simple.
Identify a color with a channel and a translation array for that color.
When Vixen gives you the value - use it as the index into the color's translation array and output that value.
You can manipulate any color or channel you like.
P. Short wrote:I'm not sure if the PIC16F688 has enough bandwidth to do 10-bit or 12-bit dimming, although I can't be sure without coding things up.
Depending on how the code is written, I think you could get 10-bit or higher raw resolution as long as the next closest used dimming level is several steps below the previous one.
don
